Glioblastoma multiforme (GBM) is the most common adult primary brain tumor. The average survival with surgery, radiotherapy, and chemotherapy is 14.6 months and is often marked by tumor recurrence after treatment. Tumor recurrence is caused by tumor cells invading through the normal brain tissue, escaping surgical resection. Current in vitro research in GBM invasion focuses on cell lines grown on plastic and in artificial collagen/matrigel matrices, and in vivo work is often studied using static, fix-and-stain techniques. These approaches either oversimplify the composition of brain tissues or neglect a critical character of GBM cells, i.e. their dynamic invasion. We propose to develop and use a new live-cell platform that will allow us to study the invasion of primary GBM cells at single-cell resolution, in live human brain tissue slices. This dynamic platform will allow us to establish the mode of migration of GBMs during invasion (mesenchymal vs. ameboid), the type of cellular protrusions generated by GBMs to invade, and the role of integrins and CD44 using shRNA-induced knockdown. Additionally, this model provides a mechanistic view of the effect on GBM invasion of cilengitide, a specific ?v?3 integrin inhibitor currently in phase III trials for GBM. Specific Aim 1. To quantify tumor single-cell invasion in brain organotypics, time-lapse fluorescence, differential interference contrast (DIC), and reflection confocal videos will be collected simultaneously to determine baseline movements and invasion characteristics of individual control GBM cells. Measurements include total distance traveled, average cell velocity, and persistence time and distance, as well as protrusion activity and dynamics, intercellular distance (time-dependent tumor cell density in the slice), and cell-induced traction forces on their surrounding microenvironment. Additional information relating tumor cell movement to extracellular structures such as vasculature and neurons will be obtained. Specific Aim 2. To determine the role of integrins and CD44 in the invasion of GBM cells in brain organotypics, we will perform knockdown and inhibition studies. Expression of integrins ?2, ?3, ?5, ?6?1, ?v?3, and ?v?5 and CD44 in GBM tumor cells will be determined using ELISA, PCR, and Western blotting techniques. Then, selective knockdown of expressed integrins ?2, ?3, ?5, ?6?1, ?v?3, and ?v?5, in the GBM tumor cells will be used to quantify changes in the cell invasion. Knocking down one integrin at a time will allow invasion changes to be attributed to the knocked down integrin. Changes in cell motility, protrusion dynamics, microenvironment deformations, and cell density of GBM cells in organotypic brain slices following treatment with cilengitide, a specific ?v?3 integrin inhibitor in phase III trial for GBMs, will be quantified and compared to control cells. The effect on cell motility, protrusion dynamics, microenvironment deformations, and cell density of GBM cells will be determined in the setting of CD44 knockdown. Finally, antibody-mediated inhibition of CD44 binding to hyaluronic acid will be used to determine the effect on cell motility, protrusion dynamics, microenvironment deformations, and cell density of GBM cells in organotypic brain slices.